1
A critical review of the formation of mono and dicarboxylated
metabolic intermediates of alkylphenol polyethoxylates during
wastewater treatment and their environmental significance
T.Y. Chiua, N. Paterakis
a, E. Cartmell
a*, M.D. Scrimshaw
b and J.N. Lester
a
a School of Applied Science, Centre for Water Science, Cranfield University, Bedfordshire, MK43 0AL,
United Kingdom.
b Institute for the Environment, Brunel University, Uxbridge, Middlesex, United Kingdom.
*Corresponding author. Tel: +44 1234 754853
Fax: +44 01234 754109
Alkylphenoxyacetic acids, the metabolic biodegradation products of alkylphenol
ethoxylates, are commonly found in wastewaters and sewage effluents. These persistent
hydrophilic derivatives possess intrinsic estrogenic activity which can mimic natural
hormones. Their concentrations increase through the sewage treatment works as a result
of biodegradation and biotransformation and when discharged, can disrupt endocrine
function in fish. These acidic metabolites represent the dominant alkylphenolic
compounds found in wastewater effluent and their presence is cause for concern as,
potentially through further biotransformation and biodegradation, they can act as
sources of nonylphenol which is toxic and estrogenic. The review aims to assess the
mechanisms of formation as well as elimination of alkylphenoxyacetic acids within
conventional sewage treatment works with the emphasis on the activated sludge process.
In addition, it evaluates the various factors influencing their degradation and formation
mailto:[email protected]
2
in both laboratory scale and full scale systems. The environmental implications of these
compounds are considered as is the need for tertiary treatment processes for their
removal.
1. INTRODUCTION
Alkylphenol ethoxylates (APEOs), introduced in the 1940s, are the second largest group
of nonionic surfactants in commerical and industrial use. They are incorporated as
additives in detergents, pesticide formulations, dispersing agents for wool scouring,
hydrogen peroxide bleaching and dyeing processes. The worldwide production of APEOs
was estimated at 500,000 tons in 1997 of which 80% were nonylphenol ethoxylates
(NPEO) and 20% were octylphenol ethoxylates (OPEO) (Renner, 1997). Production of
APEOs in Western Europe has declined between 2000 to 2002 from 116,000 to 83,000
tons respectively (Cefic, 2002). The usage of NPEO in the U.S. was approximately
130,600 tons in 2006 (ICIS Chemical Business Americas, 2007) whilst the consumption
of these compounds in the growing Asian economies is expected to rise including an
increased demand for nonylphenol (NP) in India. The Indian Union Commerce
Ministry’s statistics show that the importation of NP (and similar chemicals and their
isomers) have grown from 818 tonnes in 2000, to 23,843 tonnes in 2006. Most, if not all,
of NP would have been used in NPEO production to contribute to the estimated annual
NPEOs consumption of between 40,000 and 44,000 tonnes in India (Dutta, 2008).
As a result of their use, these commerical APEOs, consisting of a complex mixture of
ethoxy homologues and alkyl isomers (European Commission, 2002, 2003; Petrovic et
3
al., 2002a; Knepper et al., 2003) are discharged to sewage treatment works (STWs) or
released directly into the aquatic environment (Gibson et al., 2005; Koh et al., 2005;
Langford et al., 2004, 2005a). After secondary wastewater treatment usually more than
95 % of the complex mixtures are degraded to various stable and persistent metabolites
such as short chain APEOs i.e. mono- and di-ethoxylates (AP1EO and AP2EO),
alkylphenols (APs), alkylphenoxyacetic acids i.e. carboxyalkylphenol
polyethoxycarboxylates (CAPECs) and alkylphenol ethoxycarboxylates (APECs) (Fig 1)
which are frequently detected in various water bodies of Europe, North America, Japan
and Asia (Table 1).
Please insert Fig 1 here
These metabolites are recognised as endocrine disruptors and unlike their parent
compounds are also toxic to both marine and freshwater species (Comber et al., 1993;
Jobling and Sumpter, 1993; McLeese et al., 1981; Purdom et al., 1994). These effects are
dose dependent, as NP can cause intersexuality at “high” concentrations (concentrations
exceeding 1 μg l–1
in in vivo studies) (Gray and Metcalfe, 1997; Johnson and Sumpter,
2001), but not at low concentrations (Nimrod and Benson, 1998; Sumpter, 2002).
As a consequence of poor degradability and toxicity of these metabolites, APEOs have
been replaced in household applications in most western countries with the exception of
the U.S. and Japan, mainly by alcohol ethoxylates (Loos et al., 2007). Throughout
northern Europe (Scandinavian countries, U.K. and Germany) a voluntary ban on APEO
4
use in household cleaning products began in 1995, and restrictions on industrial cleaning
applications in 2000 (Renner et al., 1997). Furthermore, the European Union passed in
2003, an amendment to directive 76/769/EC; the marketing and use directive for NPEOs
which was implemented into national laws of each EU member country by 2005. The
directive states that NP and NPEO “may not be placed on the market or used as a
substance or constituent of preparations in concentrations equal or higher than 0.1 % by
mass”. This directive applies to most of the industrial applications including many of its
uses in industrial cleaning, textile industries and in metal working. However, mainly
because of lower production costs, APEOs are still being used in southern European
countries in substantial amounts in institutional and industrial applications. There are no
regulatory restrictions on the manufacture, processing or use of NPEOs in the U.S.
(APERC, 2004). Some efforts have been implemented by the U.S. Environment
Protection Agency to encourage and recognise companies, facilities and others who
voluntarily phase out or commit to phasing out the manufacture or use of NPEOs under
the Safer Detergents Stewardship Initiative which is designed to protect aquatic life.
Despite the legal and voluntary ban on APEOs in the EU, these compounds are still found
in some wastewater effluents, especially in industrialised regions. This is because the use
of APEOs in some industrial applications is not restricted. In Catalonia, a heavily
industrialized area in the northeast of Spain, industries (i.e. leather tanning, textile, pulp
and paper industries) use these surfactants in substantial amounts and as a result they are
discharged into the municipal sewer systems to be treated in municipal STW together
with urban wastewaters (Petrovic et al., 2007). In these effluents, APECs were found to
5
be the dominant metabolites at 250 μg l–1
. The dominance of these carboxylic metabolites
including CAPECs is also observed in various aquatic environments i.e. in biologically
treated wastewater, receiving rivers and the marine environment (Ahel et al., 1994b;
Field and Reed, 1996; Marcomini et al., 1990; Ding and Chen, 1999; Ding and Tzing,
1998; Ding et al., 1999; Gross et al., 2004). As much as 66% and 63% of the total APEO
metabolites detected in Italian sewage effluents were APECs and CAPECs (Di Corcia et
al., 1998a, 2000). Table 1 summarises the concentrations of these metabolites found by
some of the major worldwide investigations.
Please insert Table 1 here.
The high concentrations of APEC and CAPEC residues detected in the aqueous phase in
environmental samples reveal that these metabolites are polar and recalcitrant to further
biotransformation (Johnson and Sumpter, 2001; Hoai et al., 2004). These properties allow
the metabolites, whose concentrations increase through STWs as a result of
biodegradation and biotransformation to remain in the effluent of STWs and resist
removal. It was observed that CAPEC residues persisted in an aliquot of effluent
inoculated with microorganisms from an activated sludge STW for more than five
months after their generation (DiCorcia et al., 1998a). Due to their recalcitrance and their
hydrophilic character, they are also among the most frequently detected contaminants by
the United States Geological Survey in surface waters (Kolpin et al., 2002).
6
Several workers have expressed concerns about these metabolites persisting in aquatic
(Shang et al., 1999) and surficial marine sediments (Ferguson et al., 2000, 2001) where
they may undergo remobilization and bioaccumulate into the food chain (Ferguson et al.,
2000). A recent study showed that nonylphenol polyethoxy carboxylates (NPECs)
contributed to 33.7 % of the levels of nonylphenolics found in the tissue of benthic
invertebrates (Mayer et al., 2007). In Japan, it has been reported that the concentrations of
NPECs in sewage effluents and rivers could be one or two orders of magnitude higher
than those of NP (Isobe and Takada, 2004) suggesting that NPECs might be more
important than NP in terms of endocrine disruption on the basis that the relative
estrogenic potency of NPECs to NP is 0.63 (Servos et al., 2000).
Furthermore, through further biotransformation or biodegradation and during wastewater
chlorination, carboxylated alkylphenoxy ethoxylates (CAPECs) can act potentially as
sources of AP, short chain APEOs and halogenated APECs (such as Br-APECs, or Cl-
APECs) which are estrogenic to fish and other aquatic organisms (Kinae et al., 1981;
Reinhard et al., 1982; Ball et al., 1989; Ahel et al., 1994a, 1994b; Jobling et al., 1996).
The formation of halogenated derivatives of the alkylphenols and acidic alkylphenols,
mostly brominated compounds, were reported at μg l–1
levels in wastewater effluent and
receiving river water after disinfection with chlorine in the presence of bromide ions in
the wastewater treatment plant (Kinae et al., 1981; Reinhard et al., 1982; Ventura et al.,
1988; Fujita et al., 2000). The toxicity and estrogenicity of halogenated APECs were
found to retain a significant affinity for the estrogen receptors in in vitro tests and their
7
acute toxicity to Daphnia magna was higher than their nonbrominated precursors APEOs
and APECs (Maki et al., 1998; Garcia-Reyero et al., 2004).
Detailed studies of these compounds are hindered by the lack of commercially available
authentic standards (Montgomery-Brown and Reinhard, 2003) and the nature of both
APECs and CAPECs. They partially coelute when using gas chromatography and their
mass spectra share many features. This is a consequence of the complex branching
pattern of the nonyl group in nonylphenol polyethoxylates which form numerous isomers
that are difficult to separate and quantify. To date, CAPEC residues have been
determined semi-quantitatively by comparing their summed selected ions with the base
ion of a specific internal standard (Ding and Chen, 1999; Ding and Tzing, 1998; Ding et
al., 1999, 1996; Montgomery-Brown et al., 2003), or through biodegradation test of
APEOs (DiCorcia et al., 1998a, 2000) and NP2EC (nonylphenoethoxy carboxylates)
(Jonkers et al., 2001). The identification and quantitation of CAPECs so far has been
uncertain and subject to inaccuracies. Due to these difficulties, few reports have appeared
concerning the abundance of both APECs and CAPECs and how APECs are transformed
into CAPECs under environmental conditions.
The relatively high concentration, environmental persistence, hydrophilic nature and the
potencies of these carboxylated APEOs suggest that APECs might be more important
than APs in terms of endocrine disruption and hence have motivated on-going efforts to
identify them. However, little is known about the occurrence and fate in the aquatic
environment of APECs and CAPECs due to lack of commercially available authentic
8
standards. This review intends to assess and elucidate the mechanisms of formation as
well as the elimination of the carboxylated alkylphenol ethoxylates within conventional
STWs with emphasis on the activated sludge process. Concomitantly it aims to extend
our knowledge by filling the void regarding their behaviour and fate within STWs, thus
providing a significant tool to the wastewater industry treating and managing this
particular class of endocrine disrupting chemicals (EDC) within STWs.
9
2. FACTORS INFLUENCING FORMATION AND DEGRADATION
2.1 Small laboratory studies
2.1.1. “Abiotic” versus “Abiological”
Light dependence NPEC formation was reported by Wang et al (2006). Negligible
concentrations of NPECs were found in the dark under particle free, sterile conditions
whilst the production of NPECs under sterile conditions in the presence of light was
24.7 nmol l–1
and 6.39 × 102 nmol l
–1 at 12 h and 120 h respectively. The abiological
production of NPECs was believed to arise from the photo-(catalytical) oxidation of
NPEOs, as many compounds present in natural waters can act as photosensitizers like
humic acids, Fe(III)-aquo complexes and nitrates (Ahel et al., 1994d). These compounds
can participate in photocatalytical reactions and have a strong oxidation ability towards
organic pollutants (Haag and Hoigne, 1985; Zepp et al., 1985; Mansour et al., 1997;
Fukushima and Tatsumi., 2001). The half life of NP in lake water exposed to a mercury
vapour lamp (intensity 10 times greater than natural sunlight) was approximately 45
minutes; in distilled water, its half life was approximately five times longer. Another
factor to be taken into consideration in the photodegradation rate of APEOs is the light
intensity. It was demonstrated that at depths of 20-25 cm, photodegradation rates were
approximately 1.5 times slower than rates observed at the surface which was reported to
be between 10 and 15 hours for NP (Ahel et al., 1994d).
In the presence of microorganisms, NPECs production increased after 12 h and 120 h to
1.06 × 102 and 8.81 × 10
2 nmol l
–1 respectively (Wang et al., 2006). The formation rates
increased by 4.3 and 1.3 times over the corresponding test periods. The increased NPECs
10
concentrations was attributed to the influence of the microorganism; as the production of
NPECs under particle free, non-sterile and sterile conditions was much less compared to
the amount of NPEOs degradation especially in non-sterile treatment. These authors
suggest that either NPEC is further degraded relatively quickly, or that other degradation
pathways exist. This is unsurprising as the NPEO degradation pathway does not
necessarily produce NPECs as shown in Fig 2. The rates of NPEC formation inferred
from the data from Wang et al (2006) where NPEC forms at a faster rate in the presence
of microorganisms compared to the rates observed due to photolysis, then biodegradation
as a breakdown mechanism is more significant than photolysis.
2.1.2. Suspended Particle Matter
In a laboratory scale study to ascertain the influence of inorganic suspended particulate
matter, Wang et al. (2006) observed decreased NPEC concentrations in the presence of
these suspended inorganic particles. The presence of suspended particles is thought to
encourage a possible shielding effect to light which limits the photo-oxidation and also
allows some NPECs to associate with the solid phase (Fukushima and Tatsumi, 2001),
although the partition coefficient values of NPECs are quite low. Additionally, they
found that the adsorption of NPEOs to the inorganic particulates hindered its
bioavailability to form NPECs. The impact of hindered bioavailability within the
biological mixed liquors in the activated sludge process arising from inorganic particulate
matter may not be significant since these particles are present in small amounts (typically
less than 10 % of the biomass).
11
2.1.3. Effect of Organic Matter
Recent surveys conducted in Japan on aquatic environments demonstrated the occurrence
of long-chain carboxylates in a relatively polluted river water, suggesting the
involvement of biochemical oxygen demand (BOD) constituents (Ito et al., 2002). These
studies indicated that BOD constituents may contribute to determining the degradation
route of APEOs. As such, Hayashi et al. (2005) investigated the effect of organic matter
such as yeast, glucose and methanol on the biodegradation of NPnEOs and their
metabolites under a modified OECD 301E biodegradation test protocol. When organic
matter was absent, small amounts of short chained NPECs were detected which arose
from the oxidation of the short chained NPEOs. Biodegradation tests in the presence of
organic matter as a carbon source using NP2EO, NP3EO, NP5EO and NP10EO as
precursors resulted in the formation of the corresponding NPEOs-carboxylates i.e.
NP2EC, NP3EC, NP5EC and NP10EC. This suggests that oxidation is independent of EO
chain length. The NP10EC formed 85 % of the original concentration (molar based) on
the 15th day. The biodegradation of NPEOs and its corresponding formation of NPECs in
the presence of low growth bacterial population led Hayashi et al. (2005) to suggest that
the NPECs may be generated by cometabolism of NPEOs when methanol is being
utilised oxidatively as the sole carbon source. Several other workers have reported that
increasing the amount of yeast extract enhanced the transformation of NP2EO to NP in
cultured medium (Liu et al., 2006) and that of NP and NP1EO in river sediment (Fujii et
al., 2000). Ushida et al. (2003) also described that a novel NP-degrading Sphingobium
amiense sp. nov. could degrade NP in the presence of an organic nutrient such as yeast
extract but could not use NP as a carbon source. An appropriate addition of
12
supplementary nutrients seems to be indispensable for promoting degradation; such
supplementation corresponds to a biostimulation technique in the bioremediation of
pollutants in situ.
2.2 Full scale studies
2.2.1. Unit treatment process type
The type of secondary treatment used at a STWs is important in determining the relative
concentrations of the NPEO parent components and their metabolites present in effluent.
Concentrations of NP1-3EO > NPEC indicate that the predominant degradation route has
been under anaerobic conditions, while if NP1-3EO < NPEC then this would indicate that
aerobic degradation is predominant (Tarrant et al., 2005, Barber et al., 2000; Petrovic and
Barcelo, 2001).
Samples taken from several municipal sewage treatment works in Switzerland were
analysed and various degrees of elimination for nonionic surfactants were observed
(Giger et al., 1987). In general, nonylphenolic compounds were removed efficiently by
the activated sludge treatment with increased carboxylated metabolite concentrations in
all plants. The STW with the highest NP3-20EOs, NP1-2EO and NP removal i.e. 99 %, 78
% and 98 % respectively showed the lowest carboxylated metabolite formation (573 %).
In contrast, the highest formation of NPECs of ca. 33200 % was accompanied by a net
formation of NP1-2EO (61%) and decreased elimination efficiencies for all nonylphenolic
compounds.
13
In the same study Giger et al (1987) analysed the concentrations of metabolites in each
step of the Uster STW. Primary treatment resulted in a reduction in the concentration of
the lipophilic NP and NP1-2EO accompanied by a small increase in NPECs i.e. 3 %. A
significant proportion of the NPECs, i.e. 58 %, was found in the secondary effluent and
in the activated sludge. The activated sludge process reduced concentrations of NP, NP1-
2EO and most significantly the NP3-20EO. Tertiary treatment had no effect on the more
hydrophilic NP1-2EC. This supports the notion that NPECs are the biodegradation
products of NPEOs. The same finding was also reported by Ahel et al (1994a) in effluent
samples taken from various STWs in Switzerland.
A comprehensive study on the occurrence, transformation and elimination of NPEO
compounds in eleven full-scale activated sludge sewage treatment works in the Glatt
Valley, Switzerland was undertaken by Ahel and co-workers (1994a). The NPnEO
oligomer distributions in the effluents preceding each treatment were significantly
different. In primary effluents, the principal components were NP3-20EOs accounting for
approximately 82 % of the total followed by NP1-2EOs (circa 12 %), 3 % NP and 3 %
NP1-2ECs. However, in the secondary effluent, NP1-2ECs were found to be the most
abundant (circa 46.1%) followed by NP3-20EOs which contributed to circa 28 % whereas
NP1-2EOs accounted for circa 22 % and 4 % NP. This 14.9 fold concentration increase
from primary to secondary effluents suggests that the biological activated sludge
treatment favours the formation of NP1-2ECs.
14
Recently, Nakada et al. (2006) carried out a comprehensive study at a sewage treatment
work in Japan. In their summer survey, NPEC removal efficiencies of 1, 33 and 31 %
were found after the primary settling tank, aeration and final settling tank and disinfection
tank respectively. This is in contrast to various workers who have reported a net increase
in NPECs in the secondary effluents compared to influents.
Bennie (1999) summarized the levels of NPECs found in Canadian STW effluents and
reported that the concentrations of NP1-2EC increased with increased degrees of treatment
in contrast to the decline in NP1-2EO concentrations. They reiterated that the nature of the
inputs and type and degree of treatment strongly influence the concentrations and relative
proportions of NPEOs released in final effluents. This is in agreement with Barber et al.
(2000). The median concentrations of NPECs in the effluents of various processes are as
follows in descending order: STW only trickling filter > activated sludge STWs without
trickling filter > activated sludge STWs with trickling filter > activated sludge STWs
with granular activated carbon treatment (Barber et al., 2000). However, these data may
be interpreted inaccurately since neither detailed mass balances nor removal efficiencies
were reported.
This finding is consistent with earlier work (Ball et al., 1989) which observed in time
cause experiments that in activated sludge OP2EC was a significant intermediate, whereas
OP2EC accumulated only to a minor extent in the crude sewage. When comparing the
two experiments, it appears that the dominant factors affecting formation and
accumulation of OPECs in aerobic environments include composition and concentration
15
of microorganisms with the higher concentration of microorganisms in activated sludge
possibly favouring the formation of OPECs.
From the various studies reported above, the removal efficiencies of the carboxylated
compounds as well as other alkylphenolic metabolites are shown to vary with treatment
process conditions. Other variables such as sludge age, temperature and hydraulic
retention time are important parameters for organic degradation (Langford et al., 2005a,
2005b) and may also play an important role which is discussed below.
2.2.2. SRT and HRT
The solid retention time or sludge age (SRT) is the mean residence time of
microorganisms in the biological reactor and only organisms that can reproduce
themselves during this time can be retained and enriched. High SRT will therefore allow
the enrichment of slowly growing bacteria and consequently the establishment of a more
diverse biocenosis with broader physiological capabilities (e.g. nitrification). It has been
recognized that the complete biodegradation of surfactants requires a consortium of
bacteria due to the limited metabolic capacities of individual species (van Ginkel., 1996;
Langford et al., 2005b). Opportunities for commensalism and synergism to occur exist in
a consortium. Such interactive effects lead to more effective biodegradation than is
possible by any individual microorganism. This condition is often realised with a
membrane bioreactor (MBR) which is a modification of the activated sludge process. In
the few studies undertaken comparing MBRs to conventional activated sludge systems,
16
the higher removal efficiencies of alkylphenolic compounds were attributed to the
extended length of SRT (Gonzalez et al., 2007; Terzic et al., 2005).
Overall NPEO eliminations of 74 % and 87 % accompanied by NPEC formations of 1471
% and 429 % were achieved in the conventional STW and MBR respectively (Terzic et
al., 2005). The NPEC contributed to 67 % and 42 % of the total nonylphenolic
composition within the secondary effluent and the MBR treated wastewater respectively.
These differences have been attributed to the longer SRT in MBRs as compared to the
conventional activated sludge STW which allowed the development of microbial
consortia capable of biotransforming more persistent oligmers (Terzic et al., 2005). In
another study, the conventional STW achieved overall elimination of nonylphenolic
compounds of around 54 % with NP1-2EC accounting for up to 60 % of the total
nonylphenolic compounds in the effluent. In the MBR system, the elimination efficiency
of nonylphenolic compounds was 94 % and the percentage proportion of NP1-2ECs in
MBR effluent was 35 %. Up to 73 % of NP1-2ECs were removed in the MBRs as
compared to a net formation of more than 70 % occurring in the conventional activated
sludge STW (Gonzalez et al., 2007). Additional tests were carried out to determine if the
decreased concentrations arose from degradation or sorption onto the sludge. These
workers confirmed that the better elimination in the MBR was a result of better
degradation rather than from sorption. The better performance of the MBR was attributed
to the higher sludge age and the lower sludge load, which gave more time for the
degradation of the organic contaminants and/or to the better adaptation of the
microorganisms. However, both studies did not provide the SRT for both processes
17
which leads us to question if SRT is really a significant factor as opposed to other
operating parameters for the different removal performances observed in both processes.
Furthermore, if NPEC is degraded dependency of SRT, the determination of a critical
SRT would not be possible.
Clara et al. (2005) evaluated the impact of SRT on treatment removal efficiencies using
conventional STWs and MBRs operated at different SRTs. The operation of STWs with
SRTs suitable for nitrogen removal (SRT >10 days) exhibited increased removal
potential for NPEOs and NPECs. Comparable results are reported by Ahel et al. (1994a)
who observed the highest removal rates for NPEOs in low-loaded conventional municipal
STWs operating at high SRTs. The NPEC productions of 1.4 % and 49 % corresponding
to SRTs of 114 and 237 days respectively were observed in the conventional STW under
other comparable operating conditions i.e. HRT, food to microorganism ratio and
temperature. Similarly, increasing the operating SRT in the MBR from 10 days to 27
days resulted in increased NPECs production of 142 % to 604 % respectively. Negligible
differences in the removal potential of the nonylphenolic compounds were observed
between the conventional STW and MBRs which led these workers to conclude that the
membrane did not allow further retention of nonylphenolic compounds and that there
were no differences in treatment efficiencies between the two treatment techniques (Clara
et al., 2005). This may not be the case for NPEC since observations were based on the
sum of the short chained metabolites i.e. NP1-2EO, NP and NP1-2EC to which these
authors reasoned that the interactions between the different fractions required an
integrated evaluation. This is debatable since other major metabolites such as the
18
dicarboxylates are not even included. Moreover, the operating parameters for both
processes were not similar to allow for a fair comparison.
From the limited literature, there is no doubt that SRT plays a role in the removal of
NPECs. However, in the studies comparing conventional STW and MBRs, the improved
removal of NPECs could not distinctly and solely be attributed to the SRT since other
additional characteristics associated with MBRs have been expected to contribute to the
enhanced removals. One of these is that MBRs have a low sludge load in terms of
biochemical oxygen demand (BOD). In this situation, the bacteria are forced to
mineralize poorly degradable organic compounds. The notion of loading has been
demonstrated indirectly (Ahel et al., 1994a; Clara et al., 2005). The NP1-2EC formations
of 240 % and 280 % were observed in low-loaded and high loaded conventional STWs
respectively (Ahel et al., 1994a).
Another important process variable to consider is the hydraulic retention time (HRT).
This is the amount of time the sewage spends in the biological reactor during activated
sludge treatment and determines the exposure time a bacteria will have to degrade or
adsorb a compound (Langford et al., 2005a). Since APECs are hydrophilic and are
weakly sorptive, it could be predicted that longer HRT results in greater removal. The
biological degradation of OP1EC was observed after 2 days by Fujita and Reinhard
(1997) after exposure to groundwater microorganisms. OP1EC was utilized as a sole
carbon source and after day 4, no OP1EC could be detected. In MBR studies, Gonzalez et
al (2007) observed higher NPECs removal efficiencies than previous workers (Terzic et
19
al., 2005) and attributed this, amongst other possible factors, to the longer HRT used in
their study, 10 hours compared to 5.7 hours (Terzic et al., 2005), which favoured a more
complete degradation of NPEC. Although not explicitly noted by Clara et al. (2005),
operating the MBR and conventional STW under similar conditions other than a change
in HRT from 4 days to 13.6 days resulted in NP1-2EC formations of 126 % to 254 %
respectively. This observation opposes the view made by Gonzalez et al (2007). The
limited studies carried out on the effects of HRT on NPEC formations hinders our
judgement to assess how significant the effect of HRT is.
2.2.3. Temperature
Generally metabolic processes increase with temperature (Lester and Edge, 2001; Birkett
and Lester, 2002). Greater removal of APEOs from effluent can be observed in summer
however, this corresponds with an accumulation of their metabolites. To date, there is
little information on APECs in these studies involving seasonal variations. If APECs are
intermediate degradation metabolites of APEOs, then temperature should be of some
relevance to the production of APECs.
Studies on the biodegradation of OPEO during the percolation of contaminated sewage
through a trickling filter in Preston, U.K. have shown degradation from 20% up to 80%
between the winter and summer seasons respectively (Mann and Reid, 1971). In the Glatt
River study (Ahel et al., 1994a) NP and NP1-2EOs concentrations were measured over a
year with lower concentrations being found during summer than in winter. In a study of
the seasonal variation of concentrations of NP, NP1EO and NP2EO in the Glatt and Thur
20
Rivers, Switzerland significantly higher levels of all components were measured in
samples taken in the winter months (Ahel et al., 2000). Similar findings were also
observed more recently by Maruyama et al. (2000). Water samples were taken from three
rivers in Tokyo, Japan. Concentrations of all nonylphenolic compounds were found to
decrease with increasing water temperature. Maruyama et al. (2000) reported that the
average chain length of NPEOs in winter was longer (5-8) than in summer (2-5)
suggesting that higher water temperature and therefore greater bacterial activity may
cause faster cleavage of ethoxylate chains.
Although collative conclusions from the previous studies (Mann and Reid, 1971; Stiff et
al., 1973; Ahel et al., 1994a) indicate that temperature has a significant effect on APEOs
degradation within full-scale and experimental wastewater treatment facilities, to date,
only one study exists demonstrating clearly the effect of temperature in a wastewater
treatment plant on the production of APECs. Nakada et al. (2006) studied the removal
efficiency of an activated sludge STW in Kanagawa, Japan, in winter (January) and
summer (July) of 2004 for NP, NPEOs and NPECs. They reported NPECs removal of
approximately 50 % in summer and an overall formation of 10 % in winter.
3. METABOLIC PATHWAYS
Despite the fact that the APEO degradation has been studied for 45 years, the
mechanisms for certain APEO metabolites remain unknown. In fact one group of
metabolites (CAPECs) was not even discovered until 1996 (Ding et al., 1996). The
inclusion of terminal oxidative metabolic pathways as the dominant pathway in aerobic
21
APEO biodegradation was only realised in 2001 (Jonker et al, 2001; Sato et al., 2001).
Furthermore, although there is no disagreement about the degradability of long chained
APEOs, experimental evidence for the formation of their oxidative metabolites is
inconsistent. The formation of mono and dicarboxylated metabolites within STWs during
biological activated sludge treatment is a result of aerobic biodegradation of APEOs
(Giger et al., 1981, 1984, 1987; Stephanou and Giger, 1982; DiCorcia et al., 1994).
However, even under strict anaerobic conditions, some researchers have observed
substantial oxidation of APEOs resulting also in the formation of APECs as intermediates
(Field and Reed, 1999; Schroder, 2001; Ferguson and Brownawell, 2003). Field and Reed
(1999) studied the occurrence of NPECs in anaerobically digested municipal and
industrial sludges. NP1–4EC concentrations were in the range 27–113 mg kg–1
, with
NP2EC the most abundant oligomer, and with ortho-to-para isomer ratios ≥ 1, which
indicated the depletion of para NPEC isomers relative to ortho isomers during anaerobic
sludge treatment. By contrast, sludge that had not undergone anaerobic treatment
contained only para isomers. It is thus timely to review the proposed pathways and
mechanisms reported in the literature.
3.1. NPECs
3.1.1. Non oxidative model followed by subsequent oxidation
The biodegradation of the EO chain in the non-oxidative model proceeds through shift of
hydroxyl group from the terminal to the penultimate carbon followed by dissociation of
the resulting hemiacetal to form shorter EO chains with liberation of ethylene glycol (Fig
2 pathway 2). This hydroxyl shift model for anaerobic biodegradation of alcohol
22
polyethoxylates has been proposed by Wagener and Schink (1988). Furthermore, John
and White (1998) reported that NPEO was biodegraded by Pseudomonas putida through
the non-oxidative hydroxyl shift under both aerobic and anaerobic conditions. Since long
chained APECs were not detected in these experiments, it was generally accepted that
APEOs are biodegraded through this model to shorter ethoxy chain APEOs residues
containing one or two ethoxylate units (Fig 2 pathway 2). This has been reported by
several workers (Maki et al., 1994; Frassinetti et al., 1996; van Ginkel, 1996; John and
White, 1998; Szymanski et al., 2001, 2003) who observed the biodegradation distribution
pattern of NPEO with time. The results suggested that breakdown occurs through the
successive exoscission of the ethoxylate chain and not by direct scission between the
second and third ethoxy groups.
Under aerobic conditions, the shortened APEOs i.e. AP1-2EOs undergoes -carboxylation
of the terminal alcoholic groups producing AP1-2ECs (Manzano et al., 1999; Ahel et al.,
1994a, 1994b; Tanghe et al., 1998) (Fig 2 pathway 1). The degradation of NP15EO in
river water resulted in the formation of NP2EO and ethylene glycol, followed by
subsequent formation of NP2EC and NP1EC; NP1EO was not detected (Manzano et al.,
1998, 1999).
3.1.2. Terminal oxidative model
The APECs formation pathway described above (Fig 2 pathway 2 followed by pathway
1) generally stood as long chained carboxylated ethoxylates were previously not detected.
However, Jonkers et al (2001) observed long chain NPECs immediately upon beginning
23
their aerobic biodegradation studies in river water and concluded that the dominant
pathway in aerobic NPEO biodegradation starts with the oxidation of the terminal alcohol
on the ethoxyl chain (Fig 2 pathway 1) prior to shortening of the ethoxylate chain. No
mechanism was given for the stepwise shortening of the long chain NPEC to short chain
NPEC which is mainly NP2EC (Fig 2 pathway 4). Since small amounts of NP2EO were
formed during the aerobic degradation, these workers postulated that NP2EO were
produced in the presence of anaerobic microenvironments. These workers did not observe
the formation of higher NPEOs to show conclusively that the primary degradation
products were carboxylate derivatives as only 19% of the initial compound was
recovered. Since mineralization of NPEO metabolites often occurs very slowly (if at all),
it seems plausible that the 81% that was unrecovered was in the form of undetected
metabolites. Recently, Zhang et al. (2007) studied the aerobic biodegradation of
behaviours of two NPEC mixtures in two microcosms according to an OECD protocol
and observed the shortening of long chained NPECs to NP2EC without NPEO and NP
formation. These authors gave two possible reasons for the absence of NPEOs: 1. NPEO
were not formed in the test; 2. NPEOs may be formed and removed (by sorption or
degradation) at similar rates.
Another terminal oxidative biodegradation pathway proposed by Sato et al. (2001, 2003)
proceeds through the oxidation of the terminal EO unit (Fig 2 pathway 1) followed by
scission of the neighbouring ether bond to form the EO chains shortened by one EO unit
with liberation of glyoxylic acid (Fig 2 pathway 3). These workers studied the
biodegradation of OPEOs in a pure culture (Pseudomonas putida S-5) under aerobic
24
conditions and observed the formation of OPnEO (n=2-8) and their corresponding
OPnEC. This observation of the presence of intermediate carboxylates is also made by
several authors (Maki et al., 1996; Maeda and Mikami, 1998; Hayashi et al., 2005).
Most of the studies where NPECs were not observed with the progressive shortening of
the ethoxylated chain are presumably due to the lack of microbial consortia required for
NPEC formation (Langford and Lester, 2003). Nguyen and Sigoillot (1997) found few
Gram-negative bacteria that are able to degrade APEO with nine–ten ethoxy groups.
Whilst some Pseudomonas strains degrade only down to four or five ethoxy groups, other
species of bacteria which are unable to degrade the long chain APEO are able to degrade
the APEO with four or five ethoxy groups down to the two ethoxy group compounds.
Some NPEO degraders, such as Pseudomonas sp. Strain 14-1 and Pseudomonas sp.
Strain TR01 degraded NPEOs to NP2EO through oxidation of the EO chain to produce
carboxylates as intermediate metabolites (Maeda and Mikami, 1998).
Please insert Fig 2 here.
3.2. CAPEs
Instead of undergoing -carboxylation of the terminal alcoholic groups as described
previously to produce AP1-2ECs, the alkyl chain degradation can occur to the short
chained APEOs. In one study of STW effluent, another class of intermediates with the
alkyl chain carboxylated (CAPEs) was identified (DiCorcia et al., 1998a) (Fig 2 pathway
5). This class of metabolites was observed simultaneously with its dicarboxylated forms
25
i.e. CAPECs as described later. By comparing the CA6P2ECs spectrum with different
spectra obtained from their test solution these authors attributed the formation of CAPEs
to the oxidative attack of the alkyl side chain of the lower APEO oligmers with one and
two ethoxyl units. The new chromatographically determined class had only the alkyl
chain oxidised and they appeared with one and two ethoxy units (CAP1-2E) with CA6P2E
being the most abundant identified isomer. The CAPEs were thought to arise from the
biotransformed species of their isomeric class of NPEOs having the characteristic
structure identified by the presence of a methyl group on the branched α-carbon of the
alkyl chain. Preferential oxidation of the alcoholic terminal of ethoxylate chain over the
highly branched alkyl chain occurs due to steric hindrance. Hence CAPEs are rarely
observed as compared to APECs and they are generated less extensively than the
branched isomers. The detected CAPE did not persist as it disappeared in the
biodegradation test solution after 170 days. However, a subsequent study carried out on
“Cobis” STW located in Rome showed negligible amounts of CAPEs were present in the
effluent. Instead CAPECs accounted for more than 63% of all NPEO breakdown
products (DiCorcia et al., 2000).
3.3. CAPECs
The dicarboxylic metabolites of APEOs i.e. CAPECs were initially detected by Ding et
al. in 1996. These metabolites form via ω- oxidation of the alkyl side chain of the short
chained APECs (Fig 3 pathway 6) where the terminal methyl group is converted into a
carboxylic acid. Subsequently, shortening of the oxidised alkyl chain proceeds with α or
β- oxidation (Fig 3, pathway 7) and this leads to carboxylation on both side chains of
26
varying lengths (Fig 3). This is evident from the results of several workers (Ding et al.,
1996; Ding and Tzing, 1998; DiCorcia et al., 1998b). Carboxylation of the alkyl side
chain of NPECs observed from the product-ion mass spectra and the comparison of
methylated and propylated CNPEC derivatives are suggestive of this. DiCorcia and co-
authors (1998a) studied the biodegradation pathways of the branched alkyl chain of
NPEOs under laboratory conditions by using the OECD screening test (OECD., 1980),
using an aliquot of effluent inoculated with microorganisms from an activated sludge
STW. They found a wide range of mono and di-carboxylated biotransformation products
of NPEOs with various degrees of the alkyl chain length and ethoxylate branching. It was
concluded that a group of COP2EC isomers were produced by the ω-oxidation
mechanism of the alkyl side chain of NP2ECs (Fig 3, pathway 6). By comparing intact
APECs with their corresponding methylated forms they found additional evidence of the
formation of other di-carboxylated degradation species i.e. CA6P2EC and CA5P2ECs.
Thorough investigations allowed them to conclude that ω/β-oxidation mechanisms,
particularly of the alkyl side chains of NP2EC isomers, led to the formation of CA6P2ECs
species (Fig 3, pathway 7).
Please insert Fig 3 here.
Amongst the dicarboxylic metabolites, carboxyalkylphenoxy monoethoxy carboxylates
(CAxP1ECs, x = 6 and 8) and carboxyalkylphenoxy diethoxy carboxylates (CAyP2ECs, y
= 6 and 8) are the two most common metabolites detected in various environmental
samples (DiCorcia et al., 1998a; 2000). These dicarboxylic metabolites are postulated to
27
originate from NPEOs whilst the less abundant metabolites of C5APECs and C7APECs
arise from OPEOs. Current evidence suggests that various CAPECs metabolites will only
form after the ethoxylate chains are shortened independent of the initial breakdown
pathway (DiCorcia et al., 1998a, 2000).
3.4. Further transformation and ultimate degradation
It is generally thought that NP1ECs is transformed to NP only under anaerobic conditions
(Ahel et al., 1994a; Minamiyama et al., 2006) and anoxic conditions (Ike et al., 2002).
Recently Liu et al. (2006) reported the formation of NP from NP2EO with NP1EC and
NP2EC as the intermediate metabolites which were further transformed to shorter NPEOs
under aerobic conditions by Ensifer sp. strain AS08 and Pseudomonas sp. strain AS90.
The pathway was through the removal of one EO unit to form shorter NPEO followed by
further oxidation of the terminal alcohol group of the EO chain to carboxylic acid via
aldehyde. This process is repeated until NP is formed. This is in agreement with a
previous study which speculated that NP2EC could be further degraded and/or oxidised to
NP1EO and/or NP1EC (Maki et al., 1996).
Under aerobic conditions, although NPEC are more resistant to biodegradation than the
longer chain NPEO, they are ultimately biodegraded (Staples et al., 2001; Environment
Canada and Health Canada, 2001). Staples and co-workers (1999) determined ultimate
biodegradability, based on CO2 formation using the OECD screening test (OECD., 1980),
of NP1-2ECs and OP1-2ECs in a laboratory study by using bacterial seed inoculum from
municipal and industrial activated sludge. Half-lives for NP1-2EC and OP1-2EC in
28
bacteria-seeded water were in the range of 18-22 and 12-18 days respectively. Using
municipal settled sewage as an inocula for the OECD tests, OP1EC and NP1EC
biodegradation reached 72 % and 59 % respectively. Mixed liqour collected from
industrial STWs as an inocula exhibited similar levels of biodegradation for OP1EC and
NP1EC of 65.3 % and 66 % respectively. Although all the compounds exceeded 60% of
theoretical CO2, none of the tested compounds demonstrated 100% degradation over 28
days. These workers found greater biodegradation with OP2ECs and NP2ECs than their
shorter chain counterparts i.e. OP1EC and NP1EC in municipal activated sludge. The
mineralization is thought to involve the breakdown of the aromatic ring at the centre of
the NPEO molecule prior to the loss of the final ethoxylate or carboxylate groups, since
NP is not observed under aerobic conditions (Jonkers et al., 2001; Tanghe et al., 1999;
Staples et al., 1999).
The CAPECs metabolites are recalcitrant to further biotransformation; they persisted in
the test liquor more than 5 months after they were generated (DiCorcia et al., 1998a).
These authors stated that the lack of further CAPECs biotransformation was not due to
the lack of microorganisms since CAP2ECs to CAP1ECs conversion was observed. The
persistence of CAPECs was also evident in studies investigating the amounts present in
river waters and effluents from STWs. DiCorcia et al (2000) found that these
dicarboxylated species were by far the most abundant metabolites found in the effluents
of five STWs accounting for ca 66 %. In a recent study involving Taiwanese waters,
CAPECs were found to be the dominant alkylphenolic compounds i.e. CAPECs (up to
94.6 μg l−1
) followed by NPECs (up to 63.6 μg l−1
) (Cheng et al., 2006b). A possible
29
explanation for the persistence of CAPECs is that these dicarboxylic acid molecules form
strong intermolecular hydrogen bonds between their acidic ends resulting in greater
stability. This phenomenon is analogous with the strong interactions between two acidic
molecules forming a dimer which acts as one single molecule (Hill et al., 1993). The
significance of this for CAPECs biodegradation is that it probably hinders or restricts
further biodegradation of the aromatic ring. Additionally, these carboxylic acids possess
the ability to resonate resulting in enhanced stabilizing effects on the RCO2- ends by the
charge-delocalization species i.e. the negative charge is equally distributed between two
oxygen atoms and this increased stability is comparable to that present in an aromatic
system (Gutsche and Pasto, 1975).
4. DISCUSSION
4.1. Environmental impact
Both APECs and CAPECs are formed by degradation of longer-chain-length APEOs
during wastewater treatment and they therefore, increase in concentration during the
course of treatment and can reach levels considerably higher than those of AP or APEOs
in final effluent. Typical concentrations of APECs and CAPECs found in the
environment are up to 931 µg l–1
and 755 µg l–1
respectively (Table 1). The
concentrations and occurrence of these APECs and CAPECs are important for three main
concerns: 1. they may contribute to the estrogenicity of the discharges either individually
or as a mixture effect; 2. they can be precursors of APs and short chained APEOs and 3.
30
formation of halogenated APECs which are more toxic during chlorination in direct water
reuse purposes.
4.1.1. Contribution to overall estrogenicity
Although APECs and CAPECs are less estrogenic than APs, there have been implications
that they contribute to the estrogenicity of the waters either individually or as a mixture
effect. In a risk assessment study, Bennie et al. (2001) estimated the total estrogenic
potency of municipal effluents in Canada. When considered alone, the concentrations of
NP would not exceed the threshold for estrogenic responses. If the potential estrogenic
effects of the NP1-2EOs are added to the effect of NP then about 15 % of the sites were
expected to exceed the threshold of 1 µg l–1
. When NP1-2ECs were also added, almost 60
% of the municipal sites exceeded the threshold.
Elsewhere, it has been suggested that NPECs may contribute to the estrogenic activity
measured at sites further downstream (5km away from the STW) (Sheahan et al., 2002).
Because NPEO, NP and steroids are more susceptible to removal by sorption, these
authors suspected that other hydrophilic chemicals may play a part. Since NP1-2ECs are
estrogenic at least in trout (Jobling and Sumpter, 1993; White et al., 1994; Jobling et al.,
1996), these authors suggest that the greater solubility of NPECs and its higher resistance
to degradation under aerobic conditions would support the suggestion that it may be
making a significant relative contribution to estrogenic activity downstream of the
Keighley STW effluent discharge in the UK.
31
The relative estrogenic potencies for comparisons in both studies were based on the
vitellogenin induction in trout hepatocytes data of Jobling and Sumpter (1993) in which
the relative potency of NP1-2EC to estradiol vitellogenin induction in trout hepatocytes
was reported as 6.3 x 10–6
. These authors also demonstrated that continuous exposure of
three weeks to approximately 0.11 μM of NP1EC (30 μg l–1
) in sewage effluent caused an
inhibition on the testicular growth of male rainbow trout (O. mykiss) (Jobling et al.,
1996). Another study undertaken over for a period of 22 days found that concentrations
of >10 μg l–1
of NP1EC produced statistically significant reduction in rainbow trout
weight (Ashfield et al., 1998). Over a 35 day period, exposure to NP1EC at 1, 10 and 30
μg l–1
resulted in significant increase in fish length and weight which indicated that
NP1EC exhibit adverse effects to rainbow trout at these levels. Recent studies
demonstrate the relative estrogenic potency of NP1EC is lower where NP1EC is reported
to be 30 times less estrogenic than NP (Dussault et al. 2005) contrary to previous report
by Jobling and Sumpter (1993). Therefore, if the lower potency is applied, fewer effluent
discharges would exceed the threshold for estrogenic mediated responses arising from the
presence of NPECs would be exceeded after 2005 (Table 1).
Given their high concentrations, CAPECs and NP2ECs could be significant if they have
endocrine disrupting potential. Pure carboxylates and CAPECs of different isomeric
structures would need to be tested for endocrine activity to produce the required data. It
has been demonstrated that only the 4-tertiary isomers of nonyl and octylphenol and their
metabolites have estrogenic potency (Routledge and Sumpter, 1997).
32
Therefore, if an assessment is made solely on the available potency data and current
observed levels, NP1EC would be perceived as a very weak contributor to the endocrine
problem. However, if their potential to form NP and other halogenated NPEC which are
more toxic and estrogenic is taken into account, then the presence of these compounds
could be significant. The concern about the ability of carboxylates to form NP is
amplified by the recent findings of vitellogenin induction in juvenile rainbow trout
following prolonged exposure to concentrations of NP as low as 1.05 μg l−1
(Ackermann
et al., 2002).
4.1.2. Precursors of APs and short chained APEOs
Recently, Loos et al (2007) analysed the levels of NP and NPECs in discharges, effluent
and river receiving waters from the textile industry and their corresponding STWs in
Belgium and Italy. These workers report higher concentrations of NP in receiving rivers
compared to the effluents and attributed this to NP being the final degradation product of
NPEO surfactants. This suggests that the carboxylated metabolites could potentially
contribute to NP levels. High levels of NP in river sediments, even at distances of up 10
km from the point of STW effluent discharges, were reported by Isobe et al. (2001) who
concluded that the water of the Sumidagawa River could be potentially hazardous to fish.
They suggested that the ubiquitously high concentrations of APs in the riverine sediments
could be attributed to the large proportion of wastewater effluent in the river water. It was
also postulated that increased NP concentrations may arise from NP1EO which is
anaerobically degraded even though these hydrophobic metabolites were released in
33
small amounts. The study failed to identify the quantities of NPECs which can also
anaerobically degrade to form NP (Minamiyama et al., 2006). The notion of NPECs
contributing to the NP levels detected in the environment has been suggested by Conn et
al. (2006). These workers attributed the apparent production of NP in wetland systems, a
predominantly mixed redox environment, to the presence of NPECs in the effluent which
was released into the system.
4.1.3. Halogenated APECs
It has been suggested that APECs should be regarded as potentially very important
groundwater contaminants at river water infiltration sites. Coupled with the
predominance of these carboxylic compounds in river water samples, the higher
solubility of APECs and their resistance to biodegradation in aerobic conditions (Ahel et
al., 1987) are possible reasons for their higher mobility in aquifers compared to APEOs.
However, it has been concluded that additional investigations are necessary to draw firm
conclusions about the behaviour of APECs during infiltration (Ahel et al., 1994c).
Another consequence of the persistence of APECs and CAPECs is that these compounds
will find their way into drinking water treatment plants. In some cases mono and
dicarboxylated metabolites have been found in groundwater and drinking water samples
(Ventura et al., 1988, 1992; Petrovic et al., 2001, 2002b). Pre-chlorination was found to
reduce the concentration of short-ethoxy chain NPECs and NPEOs by 25–35% and this
arises partly from their transformation to halogenated derivatives (Petrovic et al., 2002b).
After pre-chlorination, halogenated nonylphenolic compounds represented approximately
34
13% of the total metabolite pool, of which 97% were in the form of brominated acidic
metabolites. Little is known about the environmental significance and toxicology of
brominated and chlorinated alkylphenolic compounds. Reinhard and co-workers (1982)
suspected that the occurrence of mutagenicity in wastewater is correlated with the
formation of brominated alkylphenolic byproducts; however, their preliminary
experiments conducted with BrAPECs failed to confirm this hypothesis. Maki et al.
(1998) determined that both BrNPEOs and BrNPECs show higher acute toxicity to
Daphnia magna than do their nonbrominated precursors NPEOs and NPECs. The 48 h 50
% lethal concentrations (LC50s) of NP2EO, BrNP2EO, NP2EC and BrNP2EC were
reported as 0.148, 0.067, 0.990 and 0.141 mg l–1
respectively. A recent study, employing
recombinant yeast assay (RYA) and enzyme linked receptor assay (ELRA) for the
determination of estrogenic and anti-estrogenic activity, showed that halogenated
compounds acted as weak estrogens when compared to NP but retained a significant
affinity for the estrogen receptors. This suggests that they may be still able to disturb the
hormone imbalance of exposed organisms as the halogenated NPECs behaved as
estrogenic antagonists in the RYA (Garcia-Reyero et al., 2004). In addition, an increased
cytotoxicity for the carboxylated derivatives in both yeast and mammalian cells was
detected. Although derivatization mask the apparent estrogenicity of nonylphenol, the
resulting compounds still represent a potential hazard since they are still able to bind the
estrogen receptor and to influence the physiological response to estrogens.
The reported levels of these halogenated compounds in the aqueous environment are
scarcely available to accurately carry out any risk assessments. However if the levels are
35
similar to those found in drinking waters, ranging from >NP>NP1EO. Acidic metabolites were completely degraded within 4
to 6 minutes (initial concentration, 0.4–1.0 mg l–1
), the NP concentrations were reduced
by 75–80% in 6 minutes, while only 25–50 % of NP1EO was eliminated in the same time.
Studies on the decomposition of NPEOs, NPECs and NP by electron beam irradiation,
36
ozone and combined ozone/ electron beam irradiation treatment in laboratory batch
conditions using spiked tap water were carried out by Petrovic et al. (2004). The results
indicated that single electron beam irradiation is the most efficient process leading to a
rapid breakdown of all nonylphenolic compounds, including NPEOs recalcitrant
metabolites, such as NP and NP1EC. The higher OH radical concentration generated by
the combined ozone/ electron beam irradiation process did not result in improved
decomposition of the compounds studied, while ozone alone achieved the lowest OH
radical production and least efficient decomposition of alkylphenolic compounds.
However, using advanced oxidation processes (AOP) to treat the carboxylic metabolites
would seem economically undesirable as the degradation efficiency of an AOP is often
limited by the radical scavenging capacity of the matrix of the treated water. In a
subsequent study carried out by Petrovic et al. (2007) using electron beam irradiation to
treat sewage effluent, NP elimination was less efficient and they attributed this to the
simultaneous elimination of NP (present in STW effluent) and its formation as
intermediate products of NPEC degradation. The complex wastewater matrix, as
indicated by the high COD and BOD values, slowed down electron beam irradiation
induced decomposition of alkylphenolic compounds and required higher doses. For
comparison, radiolytic decomposition of 500 μg l−1
of long ethoxy chain NPEOs in
spiked tap water required a 1 kGy dose (Petrovic et al., 2004); for naphthalene disulfonic
acids, a 96% degradation yield at 10 μg l−1
in a tap water was observed at 2 kGy
(Gehringer et al., 2005). For complex industrial effluents, the necessary dose to remove
90% of the most organic compounds and more than 70% of their toxicity (determined by
Vibrio fisheri Microtox test and Microcrustacean Daphnia similes test) was found to be
37
20 kGy (Duarte et al., 2002; Moraes et al., 2004). This means that the concentration of
dissolved organic carbon will play an important role in the sufficient degradation of the
NPECs from wastewater and that economic considerations have to underpin the
feasibility of the process for wastewater treatment. The other drawbacks of using AOP to
treat alkyphenolic surfactants and their metabolites in water and wastewater have been
reviewed by Ikehata and El-Din (2004).
A further method of removing carboxylic metabolites from aqueous streams involves the
use of physical separation such as membranes. Nanofiltration/reverse osmosis (NF/RO)
membrane filtration processes have all been recognized as important technologies which
can be used to remove a wide range of contaminants including endocrine disruptors. The
adsorption and release process of several endocrine-disrupting chemicals (EDCs) during
NF/RO filtration processes was examined by Nghiem and Schäfer (2006). Results
indicated that the membrane can serve as a large reservoir for EDCs. Their release may
be possible during membrane cleaning or as a result of erratic pH variation during
operation. Treatment of membrane cleaning solution should be considered carefully when
EDCs are amongst the target contaminants in NF/RO membrane filtration. Other workers
demonstrated that the physicochemical properties of the membranes and the solutions
play vital roles in their rejection efficiencies (Chiu et al., 2006). Lower rejection
efficiencies were found in experiments that employed low concentrations of compounds
(Kimura et al., 2003). This is applicable also to membranes within membrane bioreactor
systems (MBRs). Other considerations to take into account are the regeneration cost of
38
membranes, as they are prone to fouling, and their downtime. Additionally, as NF/RO
serves to improve the effluent quality, these compounds would concentrate within the
reject stream and thus careful consideration and management of the reject stream are
essential.
Considering the drawbacks of using physical and chemical methods as possible tertiary
treatment, it seems that more efforts should be placed on optimising operating parameters
of biological treatment plants. The use of enhanced biological treatment is an option
worth considering. Enhanced biological treatment such as biological nutrient removal
(BNR) systems possess some of the conditions required for the removal of these
carboxylated metabolites such as the increased SRT or HRT as described previously.
Under aerobic and anaerobic conditions, the breakdown of alkyphenolic carboxylates will
be encouraged. However concern for the presence of the breakdown end-product NP
arises in BNR systems. Anaerobic degradation of carboxylic metabolites leads to the
formation of NP which is more toxic. Studies on the anaerobic biodegradation of NP
indicate that sulfate-reducing bacteria constitute a major microbial component in the
anaerobic degradation of NP, but that the methanogen and eubacteria microbial
populations are also involved. This result was similar to that reported by Chang et al.
(2004), who studied the anaerobic degradation of NP in river sediment, and supports the
idea of using anaerobic microbes to remove NP from sludge. However the degradation
rate declined when the concentration of NP increased. This has been attributed to the
increased levels of toxicity at higher NP concentrations.
39
The ideal way to remove carboxylates from wastewater would be to oxidise them
completely. Two ways may be feasible. Firstly, to increase the aerobic section of the
wastewater treatment plant to increase the exposure time of these metabolites to aerobic
conditions. Secondly, to use strong oxidation conditions as used in industrial wastewater
treatment processes using tertiary non-biological processes. Ultimately, the question
arises at what environmental costs, in terms of energy consumption and CO2 release, will
this be accomplished at and will the solution cause greater detrimental effects to the
environment than APEOs themselves.
4.3. Environmental and economic analysis
The Environment Agency of England and Wales has instigated recently a
“Demonstration Programme” which includes evaluating tertiary treatment methods for
the removal of endocrine disrupting chemicals including APEOs. The economical and
environmental costs of proposed tertiary treatments have been evaluated recently (Jones
et al., 2007). Table 2 shows the ranges of the equipment operational costs for each of the
process options which were designated by the Environment Agency of England and
Wales.
Please insert Table 2 here.
Table 2 includes the cost of utilizing drinking water technologies to treat urban
wastewater and demonstrates that it is likely to be costly since the additional capital
expense required is almost as much as the total capital cost of a standard plant. The
capital cost of the sand filter and membranes exceeded the cost of the basic activated
40
sludge (AS) plant by £1.5 million. The potential operating costs of the extra treatment
processes are also significantly higher than standard treatment, with treatment via MF
and RO being more expensive than GAC (granular activated carbon) and O3 (ozone)
(Table 2). There is however, an economy of scale in the cost per m3 of sewage treated via
the advanced treatments, as cost was found to decrease as the size of the plant increased.
An important factor to consider when utilizing advanced treatment technologies is the
inevitable environmentally undesirable increase in energy consumption. At present this
demand would be met mainly from non-renewable sources. An energy intensive tertiary
process plant would, contribute therefore, a large amount of CO2 to the atmosphere, with
associated ramifications for global warming and climate change. In addition, there would
be increased sludge production associated with advanced treatment which would have to
be safely disposed of, ideally in an environmentally sustainable manner. At present it is
not clear how this could be achieved. Taken together these issues can increase
significantly the economic (and environmental) cost of this type of treatment (Jones et al.,
2007).
Studies attempting to calculate the environmental benefits of removing endocrine
disrupting compounds from the wastewater stream have been commissioned (AEA
Technology Environment, 2004). The report was based on introducing various
technologies to reduce the levels of these compounds in 48 of Yorkshire Water’s
wastewater treatment works in the Yorkshire Water region, England with an estimated
population equivalent of 4 million. To comply with the Water Framework Directive
41
2000/60/EC (European Commission, 2000), it is reported that additional electrical energy
of 35 GWh would be required per annum for pumping and producing ozone and UV
light. The electricity use alone would lead to an annual release of around 175 and 1
kilotonnes of CO2 and SO2 (combustion of fossil fuel) respectively.
A paradigm of wastewater treatment is that increasing effluent quality can only be
environmentally beneficial; this should be carefully considered as the benefits of
improved effluent quality are often outweighed by the negative effects on the wider
environment when process construction, operation and the increased sludge production
from the advanced treatment technologies are taken into account. Using advanced
treatment methods to treat sewage is likely to reduce pollution, but will also incur large
financial and environmental costs (Jones et al., 2007).
Before upgrading existing sewage treatment works with advanced treatment technologies,
potential alternative options need to be considered, including preventive and remediation
measures. Known dangerous chemicals should be regulated and their production
minimised, avoiding contamination of municipal sewage and aquatic environment. New
chemicals should be tested rigorously with standard methods for reliable and comparable
results. Looking at the cost of removing such compounds, it seems a better option to
eliminate the problem at the source i.e. limiting APEO production or to use substitutes
rather than to subsequently treat them.
5. CONCLUSIONS
42
The current NPEO production and consumption in EU has declined, however, in
other parts of the world notably in Asia and the U.S. substantial levels of NPEOs
are still in use.
High levels of persistent alkylphenolic carboxylated derivatives are detected in
sewage effluents.
The hydrophilic nature of these compounds means they are highly mobile and can
enter the aquatic environment in significant concentrations.
Although weakly estrogenic, the presence of these carboxylates can contribute
significantly to the overall estrogenicity found in the environment as ad mixture
effect.
Under appropriate conditions, these hydrophilic compounds can further transform
to form NP which is toxic and estrogenic.
Tertiary treatment options seem able to remove these compounds but at
undesirable economic and environmental costs.
Improving biological treatment processes and optimising their operating
conditions (i.e. SRT, HRT) may hold the key to increase the removal of these
metabolites.
REFERENCES
Ackermann G.E., Schwaiger J., Negele R.D. and Fent K. Effects of long-term
nonylphenol exposure on gonadal development and biomarkers of estrogenicity in
juvenile rainbow trout (Oncorhynchus mykiss). Aquat. Toxicol. 60, 203–221, 2002.
43
AEA Technology Environment., The Environmental and Economic Costs and Social
Consequences of Reducing Levels of Dangerous Substances Discharged from Waste
Water Treatment Works in the Yorkshire Water Region. A report produced for Yorkshire
Water, 1-10, 2004.
Ahel M., Conrad T. and Giger W. Persistent organic chemicals in sewage effluents 3.
Determinations of nonylphenoxy carboxylic acids by high-resolution gas-
chromatography mass/spectrometry and high-performance liquid chromatography.
Environ. Sci. Technol., 21, 697-703, 1987.
Ahel M., Giger W. and Koch M. Behaviour of alkylphenol polyethoxylate surfactants in
the aquatic environment – I Occurrence and transformation in sewage treatment. Water.
Res., 28, 1131-1142, 1994a.
Ahel M., Giger W. and Schaffner C. Behaviour of alkylphenol polyethoxylate surfactants
in the aquatic environment – II Occurrence and transformation in rivers. Water. Res., 28,
1143-1152, 1994b.
Ahel M., Hrsak D. and Giger W. Aerobic transformation of short-chain alkylphenol
polyethoxylates by mixed bacterial cultures. Arc. Environ. Contam. Toxicol., 26, 540-
548, 1994c.
44
Ahel M., Schaffner C. and Giger W. Behaviour of alkylphenol polyethoxylate surfactants
in the aquatic environment - III. Occurrence and elimination of their persistent
metabolites during infiltration of river water to groundwater. Water Res., 30, 37-46, 1996.
Ahel M., Scully F.E., Hoigne J. and Giger W. Photochemical degradation of nonylphenol
and nonylphenol polyethoxylates in natural waters. Chemosphere., 28, 1361-1368,
1994d.
Ahel M., Giger W., Molnar E. and Ibric S. Determination of nonylphenol polyethoxylates
and their lipophilic metabolites in sewage effluents by normal-phase high-performance
liquid chromatography and fluorescence detection. Croatica Chem. Acta., 73, 209–227,
2000.
APERC, APE Research Council, Regulatory Status of Alkylphenols and their
Ethoxylates in the United States, Canada and the European Union — November 2004.
Ashfield LA, Pottinger TG and Sumpter JP. Exposure of female juvenile rainbow trout
to alkylphenolic compounds results in modification to growth and ovosomatic index.
Environ. Toxicol. Chem. 17, 679–686, 1988.
Ball H.A. and Reinhard M. Discharge of halogenated octylphenol polyethoxylate
residues in a chlorinated secondary effluent. In: Jolley R.L. et al. (ed), Water
Chlorination: Chemistry, Environmental Impact and Health Effects, vol 5. Lewis
Publishers, Chelsea, Michigan, USA, 1505-1514, 1985.
45
Ball H.A, Reinhard M and McCarty P.L. Biotransformation of halogenated and
nonhalogenated octylphenol polyethoxylate residues under aerobic and anaerobic
conditions. Environ. Sci. Technol,. 23, 951-961, 1989.
Barber L.B., Brown G.K. and Zaugg S.D. Potential endocrine disrupting organic
chemicals in treated municipal wastewater and river water. Analysis of Environmental
Endocrine Disruptors ACS Symposium Series, 747, 97-124, 2000.
Bennie T.D. Review of the environmental occurrence of alkylphenols and alkylphenol
ethoxylates. Water Qual. Res. J. Canada., 34, 79-122, 1999.
Bennie T.D., Burnison K., Cureton P., Davidson N., Rawn T. and Servos M.R.
Uncertainties Associated with Assessing the Risk of an Endocrine Active Substance in
the Canadian Environment. Water Qual. Res. J. Canada., 36,319-330, 2001.
Birkett, J.W. and Lester, J.N. Endocrine Disrupters in Wastewater and Sludge Treatment
Processes. CRC Press Inc., Boca Raton, U.S.A., 2002.
Cefic: European Chemical Industry Council, Chemistry Sectors: Detergents, CESIO
surfactants statistics for Western Europe 2002 (2002).
46
Chang B.V., Yu C.H. and Yuan S.Y. Degradation of nonylphenol by anaerobic
microorganisms from river sediment. Chemosphere., 55, 493-500, 2004.
Cheng C.Y., Wu C.Y., Wang C.H. and Ding W.H. Determination and distribution
characteristics of degradation products of nonylphenol polyethoxylates in the rivers of
Taiwan. Chemosphere., 65, 2275-2281, 2006a.
Cheng C.Y., Li W.R., Chang J.W., Wu H.C. and Ding W.H. Synthesis and determination
of dicarboxylic degradation products of nonylphenol polyethoxylates by gas
chromatography–mass spectrometry. J. Chromatogr. A., 1127, 246–253, 2006b.
Chiu, T.Y., Lara Dominguez, M.V. and James, A.E. Critical flux and rejection behaviour
of non-circular-channelled membranes. Influence of some operating conditions. Sep.
Purif. Technol., 50, 212-219, 2006.
Clara M., Strenn B., Gans O., Martinez E., Kreuzinger N. and Kroiss H. Removal of
selected pharmaceuticals, fragrances and endocrine disrupting compounds in a membrane
bioreactor and conventional wastewater treatment plants. Water Res., 39, 4797-4807,
2005.
Comber M.H.I., Williams T.D and Stewart K.M. The effects of nonylphenol on Daphnia
magna. Water Res., 27, 273–276, 1993.
47
Conn K.E., Barber L.B., Brown G.K. and Siegrist R.L. Occurrence and fate of organic
contaminants during onsite wastewater treatment. Environ. Sci. Technol., 40, 7358-7366,
2006.
DiCorcia A., Samperi R. and Marcomini E. Monitoring aromatic surfactants and their
biodegradation intermediates in raw and treated sewages by solid-phase extraction and
liquid chromatography. Environ. Sci. Technol., 28, 850–858, 1994.
DiCorcia A., Constantino A., Crescenzi C., Marinoni E. and Samperi R. Characterization
of recalcitrant intermediates from biotransformation of the branched alkyl side chain of
nonylphenol ethoxylate surfactants. Environ. Sci. Technol., 32, 2401–2409, 1998a.
DiCorcia A., Crescenzi C., Marcomini E. and Samperi R. Liquid chromatography-
electrospray-mass spectrometry as a valuable tool for characterizing biodegradation
intermediates of branched alcohol ethoxylate surfactants. Environ. Sci. Technol., 32, 711-
718, 1998b.
DiCorcia A., Cavallo R., Crescenzi C. and Nazzari M. Occurrence and abundance of
dicarboxylated metabolites of nonylphenol polyethoxylate surfactants in treated sewages.
Environ. Sci. Technol., 34, 3914-3919, 2000.
48
Ding W.H. and Chen C.T. Analysis of nonylphenol polyethoxycarboxylates and their
related metabolites by on-line derivatization and ion-trap gas chromatography-mass
spectrometry. J. Chromatogr. A., 862, 113-120, 1999.
Ding W.H. and Tzing S.H. Analysis of nonylphenol polyethoxylates and their
degradation products in river water and sewage effluent by gas chromatography-ion trap
(tandem) mass spectrometry with electron impact and chemical ionization. J.
Chromatogr. A., 824, 79-90, 1998.
Ding W.H., Fujita Y., Aeschimann R. and Reinhard M. Identification of organic residues
in tertiary effluents by GC/EI-MS, GC-CI-MS and GC-TSQ-MS. Fresenius J. Anal.
Chem., 354, 48–55, 1996.
Ding W.H., Tzing S.H. and Lo J.H. Occurrence and concentrations of aromatic
surfactants and their degradation products in river waters of Taiwan. Chemosphere, 38,
2597-2606, 1999.
Duarte C.L., Sampa M.H.O., Rela P.R., Oikawa H. and Silveira G.G. Advanced
oxidation process by electron-beam irradiation-induced decomposition of pollutants in
industrial effluents. Radiat. Phys. Chem., 63, 647–651, 2002.
49
Dussault E.B., Sherry J.P., Lee H.B., Burnison B.K, Bennie D.T., and Servos M.R. In
vivo estrogenicity of nonylphenol and its ethoxylates in the Canadian environment. Hum.
Ecol. Risk Assess., 11, 353-364, 2005.
Dutta A.P., Authorities clueless: Carcinogen NP in a large number of products in India.
Down To Earth, 15 January 2008.
Environment Canada and Health Canada. Nonylphenol and its ethoxylates. Priority
Substances List Assessment Report. Report no. EN40-215-/57E, 98, Available from:
http://www.hc-sc.gc.ca/ewh-semt/alt_formats/hecs-sesc/pdf/pubs/contaminants/psl2-
lsp2/nonylphenol/nonylphenol_e.pdf, 2001.
European Commission, Water Framework Directive (2000/60/EC). Establishing a
Framwork for Community Action in the Field of Water Policy, 2000.
European Commission, European Chemicals Bureau. 4-Nonylphenol (branched) and
nonylphenol: European Union risk assessment report, EUR 20387 EN, Available from:
http://ecbjrcit/existingchemicals/, 2002.
European Commission. Integrated Pollution Prevention and Control (IPPC), Reference
document on best available techniques for the textile industry. Available from:
http://www.jrc.es/pub/english.cgi/0/ 733169, 2003.
50
Ferguson P.L. and Brownawell B.J. Degradation of nonylphenol ethoxylates in estuarine
sediment under aerobic and anaerobic conditions. Environ. Toxicol. Chem., 22, 1189–
1199, 2003.
Ferguson P.L., Iden R.C. and Brownawell J.B. Analysis of alkylphenol ethoxylate
metabolites in the aquatic environment using liquid chromatography-electrospray mass
spectrometry. Environ. Sci. Technol., 72, 4322-4330, 2000.
Ferguson P.L., Iden R.C. and Brownawell J.B. Distribution and fate of neutral
alkylphenol ethoxylate metabolites in a sewage-impacted urban estuary. Environ. Sci.
Technol., 35, 2428-2435, 2001.
Field J.A. and Reed R.L. Nonylphenol polyethoxylate carboxylate metabolites of non-
ionic surfactants in U.S. paper mill effluents, municipal sewage treatment plant effluents,
and river waters. Environ. Sci. Technol., 30, 3544-3550, 1996.
Field J.A. and Reed R.L. Subcritical (Hot) water/ethanol extraction of nonylphenol
polyethoxy carboxylates from industrial and municipal sludges. Environ. Sci. Technol.,
33, 2782-2787, 1999.
Frassinetti S., Isoppo A.L, Corti A. and Allini G.V. Bacterial attack of nonionic aromatic
surfactants: Comparison of degradative capabilities of new isolates from nonylphenol
polyethoxylate polluted waste waters. Environ. Techol., 17, 199-213, 1996.
51
Fujii K, Urano N, Ushio H, Satomi M, Iida H, Ushio-Sata N and Kimura S. Profile of a
nonylphenol-degrading microflora and its potential for bioremedial applications. J
Biochem., 128, 909-916, 2000.
Fujita Y. and Reinhard M. Identification of metabolites from the biological
transformation of the nonionic surfactant residue octylphenoxy acid and its brominated
analog. Environ. Sci. Technol., 31, 1518-1524, 1997.
Fujita M., Ike M., Mori K., Kaku H., Sakaguchi Y., Asano M., Maki H. and Hishihara T.
Behaviour of nonylphenol ethoxylates in sewage treatment plants in Japan –
Biotransformation and ecotoxicity. Water Sci. Technol., 42, 23-30, 2000.
Fukushima M. and Tatsumi K. Degradation pathways of pentachlorophenol by photo-
fenton systems in the presence of iron(III), humic acid, and hydrogen peroxide. Environ.
Sci. Technol., 35, 1771–8, 2001.
Garcia-Reyero N., Requena V., Petrovic M., Fischer B., Hansen P., Diaz A., Ventura F.,
Barcelo D. and Pina B.. Estrogenic potential of halogenated derivatives of nonylphenol
ethoxylates and carboxylates. Environ. Toxicol. Chem., 23, 705–711, 2004.
Gehringer, P., Eschweiler, H., Weiss, S. and Reemtsma, T. Decomposition of aqueous
naphthalene 1,5-disulfonic acid by means of oxidation processes. In: Proceedings of the
52
17th Ozone World Congress, 22–25, August 2005, Strasbourg, France, Ozone, 17th
World Congress, International Ozone Association, 2005.
Gibson, R., Wang, M.J., Padgett, E. and Beck, A.J. Analysis of 4-nonylphenols,
phthalates, and polychlorinated biphenyls in soils and biosolids. Chemosphere, 61, 1336–
1344, 2005.
Giger W, Stephanou E. and Schaffner C. Persistent organic chemicals in sewage
effluents. I. Identifications of nonylphenols and nonylphenolethoxylates by glass
capillary gas chromatography/mass spectrometry. Chemosphere, 10, 1253-1263, 1981.
Giger W, Brunner P.H. and Schaffner C. 4-Nonylphenol in sewage sludge: Accumulation
of toxic metabolites from non-ionic surfactants. Science, 225, 623-625, 1984.
Giger W., Ahel M., Koch M., Laubscher H.U, Schaffner C. and Schneider J. Behaviour
of alkylphenol polyethoxylate surfactants and of nitrilotriacetate in sewage treatment.
Water Sci. Technol., 19, 449-460, 1987.
on lez S., Petrovic M. and arcel D. Removal of a broad range of surfactants from
municipal wastewater-Comparison between membrane bioreactor and conventional
activated sludge treatment. Chemosphere, 67, 335-343, 2007.
53
Gray M.A. and Metcalfe C.D. Induction of testis-ova in Japanese medaka (Oryzias
latipes) exposed to p-nonylphenol. Environ. Toxicol. Chem., 16, 1082–1086, 1997.
Gross B., Montgomery-Brown J., Naumann A. and Reinhard M. Occurrence and fate of
pharmaceuticals and alkylphenol ethoxylate metabolites in an effluent-dominated river
and wetland. Environ. Toxicol. Chem., 23, 2074-2083, 2004.
Gutsche C.D. and Pasto J.D. Fundamentals of Organic Chemistry. Prentice-Hall, London,
1975.
Haag W.R. and Hoigne J. Photo-sensitized oxidation in natural water via OH radicals.
Chemosphere, 12, 1659–1671, 1985.
Hayashi S., Saito S., Kim J.H., Nishimura O. and Sudo R. Aerobic biodegradation
behavior of nonylphenol polyethoxylates and their metabolites in the presence of organic
matter. Environ. Sci. Technol., 39, 5626–5633, 2005.
Hill W.J., Baum J.S. and Feigl M.D. Chemistry and Life: An Introduction to General,
Organic, and Biological Chemistry. 4th
Ed., Prentice-Hall Inc., New Jersey, 1993.
Hoai P.M., Tsunoi S., Ike M., Inui N., Tanaka M. and Fujita M. Dicarboxylic degradation
products of nonylphenol polyethoxylates: Synthesis and identification by gas
54
chromatography-mass spectrometry using electron and chemical ionization modes. J.
Chromatogr. A., 1061, 115-121, 2004.
Hoai P.M., Tsunoi S., Ike M, Sei K., Lu X., Tanaka M. and Fujita M. Dicarboxylic
degradation products of nonylphenol polyethoxylates determination and structural
elucidation in water sam